A common approach for dealing with protest zeros in Contingent Valuation is dropping all answers that are exactly zero. $[$Choose all statements that are true$]$

Group of answer choices

Protest zeros can be avoided with ex$-$post questions about the rationale behind the answer

If individuals choose to reply with protest zeros at random, this approach yields a good approximation of the true aggregate WTP

The problem with this approach is that aggregate WTP may overestimate the true amount since true zeros may be dropped in the process

The problem with this approach is that aggregate WTP may underestimate the true amount since protest zeros may come from people that care a lot about the good